As is well known, system failures cause many problems to the users of the system, especially when data is lost or corrupted. Therefore, when a data processing system fails, it is important to gather information which can aid in isolating and determining the problem associated with the failure. Previously, when failures have occurred, services provided by an operating system of a central processing complex have been used to record the contents of central processing complex storage on external media in order to facilitate problem determination.
Additionally, functions have been made available with existing operating systems to record data resident in attached storage facilities to aid in problem determination activities. This data was accessed during operating system services to capture information for failures or was requested independently from failure situations.
Prior systems require software executing at the central processing complex to direct the capture of information in storage at the central processing complex or in storage at an externally attached storage facility and to provide for the central processing complex or external storage to contain the captured data. Prior systems do not allow the capturing of data at a shared facility.
Therefore, a need exists for a technique for providing diagnostic retrieval of shared facility objects which are architecturally observable by programs accessing the shared facility. A further need exists for a technique for creating a storage dump in a coupling facility. A yet further need exists for a technique for creating a storage dump which is a logical representation of the information dumped. A further need exists for a mechanism for capturing information associated with a failure the first time the failure occurs. Another need exists for a technique for capturing a coherent and atomic snapshot of structure information. A yet further need exists for a mechanism for capturing information which minimizes disruption to programs accessing the shared facility.